Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeInferential statistics
Posted by Potsdam_EM_Group on 27 Apr 2007 at 11:45 GMT
Given Figure 2C, linear regressions are not necessary. The paper generally relies on descriptive data presentation, which is fine since most of the effects are obvious from the figures alone. It appears inappropriate to conduct regressions, given (1) low statistical power and (2) arguing a null result. In some places, however, the manuscript would have benefited from a quantitative analysis of intra-individual reliability of the biases.
RE: Inferential statistics
OliviaCarter replied to Potsdam_EM_Group on 05 May 2007 at 02:32 GMT
The use of the regression analysis, in this case, was motivated by a reviewer query about our claim that the biases at onset were different to those during sustained viewing. The reviewer wanted to know if this claim was statistically valid. In order to address this issue, we ran the appropriate analysis. We agree, however, that the low statistical power in our study makes such analysis less than ideal.
Regarding the second point, we felt that our representation of the entire time course of perceptual states was the most honest and informative means of presentation. We did not believe that the addition of p values would add any further clarity to the visual representation.